Fractured gate valve

[forensiclab]

I have a fractured gate valve, with the following standards cast on the body.
PN16, BS5163, DN150 There are the letters HN, which I do not understand.
The fracture is shown in the attached photo. BS5163 indicates the valve can be made of grey cast iron, and we have confirmed it is made of this material.
I am bothered by the variable wall thickness, which on one side is 12.1mm while on the other side it is as little as 6.4mm. I cannot find minimum wall thickness specified in BS5163 and have no access to the other standards.
Can anyone assist? Is there a wall thickness problem with this valve?

 

[forensiclab]

Sorry, photo is attached to this post

 

[metengr]

forensiclab;
Yes, there is a significant wall thickness variation with this casting based on your attached photograph. Not sure but am I viewing the fracture surface shown in the photo at the perimeter or is this a cut line to expose the bore of the valve body and variation in wall thickness?

How did the valve fail?

 

[stanweld]

There is considerable rusting of the fracture surface. Why? There also appears to be some possible damage (it's not as round) near the 4:00 position in the thinned area.

 

[ginsoakedboy]

HN followed by a string of numerals probably refers to the heat number. You can look for documentation relating to that heat number from the foundry.

 

[Artisi]

Amazing what a google search will tell you.

http://www.made-in-china.com/showro...hina-Non-Rising-Stem-Gate-Valve-BS5163-.html.

What you are looking I would think is a cheaply made valve of suspect quality - hence the poor control on core placement in the pattern box.

The PN16 is probably pressure rating or similar and the DN150 is the flange size.

 

[Artisi]

PN16 is also a flange size.

 

[forensiclab]

Thank you for your comments.
To answer Metengr, you are looking at the fracture surface.
To answer Stanweld, the fracture surface is only lightly rusted. Corrosion is not a factor in causation.

I have added another photo for interest.

We concur with Artisi's comment that this is a cheap casting with poor control over core placement, leading to variable wall thickness. No one would deliberately have the wall twice at thick on one side as the other.

Our question is essentially: When does variation in wall thickness become so bad that it constitutes a defect? Is there a standard or reference that we can refer to, or do we have calculate the minimum required wall thickness, taking into account all the factors?

 

[Artisi]

" Our question is essentially: When does variation in wall thickness become so bad that it constitutes a defect? Is there a standard or reference that we can refer to, or do we have calculate the minimum required wall thickness, taking into account all the factors?"

I think you can assume that a casting is defect when the wall thickness is less than design.

 

[metengr]

Obviously, unless you have some type of recognized valve specification to procure the castings anything can be supplied. The requirements for what the minimum wall thickness, casting quality and other minimum requirements would be in the code of construction standard for this valve. There you will find your answer.

 

[MJCronin]

Hmmmmmm......

Defective and poorly made cast iron valve body....

and from China you say.........

What a total surprise.......

Ill bet we have never discussed such a thing before in this forum.

-MJC

 

[svi]

forensiclab,

ASME B16.1 for flanged fittings of gray cast iron may be a suitable guide. For a DN 150 class 125 will be 14.3 mm. It allows a 12.5 % under tolerance to this.

Can it possibly be that the casting has this min thickness at the thinner portion? The valve may have split due to an improper mount that may have subject it to a moment load, leading to the failure?

Sajit

 

[mayda]

forensiclab,

DN 150 means a nominal pipe diameter (Diametre Nominale??) 150 mm. Actual OD will be 168.3 mm = 6-5/78 = 6" nominal diameter. PN 16 is nominal pressure rating, a bit below ASME Class 150 ("150 lb"). The 16 stands for 16 kg/cm^2, and, e.g. to the 1976 DIN2401 standard (I do not have newer), for St37-2 carbon steel (similar to A106B seamless pipe / A105N forging / A216 WCB cast), allowable working pressure at 0 - 120°C was 16 kg/cm^2, reducing to 13 at 200C, and so on. 16 kg/cm^2 x 14.22 = 227 psig. Allowable working pressure for A105N flanges to B16.5, at 100°F, is 285 psig, so the pressure rating for PN 16 is a bit less.

Dimensions of the PN16 flange will be slightly different than ASME Class 150.

You will need to either get a copy of BS5163 (plus any related BS standards, as required), or find someone who can look up the required minimum thickness for the valve body. FYI, BS6153 has been replaced by BE EN 1074-1:2000, which is identical to International Standard EN 1074-1:2000. You can order online at

http://www.bsi-global.com/en/Shop/Publication-Detail/?pid=000000000030127143

As a quick comparison, the required minimum body wall thickness for a 6" nominal diamter cast carbon steel Class 150 valve to API600, is 0.47 inches = 11.94 mm. I would expect the PN 16 requirement to BS5163 would be similar, perhaps slightly thinner, but it looks like the 6.4 mm thickness is significantly undersized, and the valve should have been rejected if the manufacurer had any kind of decent quality control.

 

[rconner]

I am curious what kind of service? To what kind of (presumeably flanged?) piping material was the valve connected? I am guessing valve was exposed (seeing the flanged joint ends), but could this be confirmed, and also with what kind of piping support and pressure/temperature etc. exposure?
In any case I am guessing a lot of conceivable causes could result in the fracture of quarter inch thick gray cast iron valve case (that does not have near the fracture toughness of a modern ductile iron or steel valve)!

 

[Artisi]

But what quality is the "gray cast iron" in this case?

 

[rconner]

If this were a 6" 200W IBBM gate valve manufactured e.g. per AWWA C500 standard (which it clearly is not based on the metric information provided), and if what was measured/provided were the original manufactured wall thicknesses, I believe it could be argued the valve did not meet that standard in this and perhaps other respects. That American Water Works standard does mention specific valve "structural defects", among which is included "failure to meet minimum wall thickness requirement".
That standard also in Section "4.4.1.1 Shell thickness"
says the following, "The body and bonnet of valves shall be made of gray iron or ductile iron and shall comply with the requirements of Sec. 4.3.1. Shell thickness
measurements taken at points diametrically opposite to each other shall, when added together and divided by 2, equal or exceed the minimum metal thicknesses given in Table 1. At no point shall shell thicknesses be more than 12.5 percent thinner than the minimum metal thicknesses shown in Table 1. No continuous area of deficient thickness shall exceed 12.5 percent of the pressure containing shell area of the casting."

It is of course otherwise kind of hard to make conclusive statements regarding "quality", not knowing exactly what the valve condition, metallurgy, nor specific quality requirements were!

 

[rconner]

[The Table 1 thickness value for a 6" valve per AWWA C500 is 0.43" (10.9 mm).]

 

[forensiclab]

We have enough answers now for this question. It would seem we have to draw an inference about the quality of our valve by reference to comparable standards. In this respect it was the quantity and variety of excellent answers that has assisted us the most.

In fairness to China, as it is being criticised so severely at the moment about milk, the valve was made in India.
Thank you
Forensiclab

 

[eyec]

In fairness to China, as it is being criticised so severely at the moment about milk, the valve was made in India.

In fairness to buyers, was it purchsed from an Indian supplier, yet made in China?



Steven C
Senior Member
ThirdPartyInspections.com

 

[chicopee]

So much for third world countries trying to imitate developed industrialized nations. When will people learn,ie, you get what you paid for.